Hackers Stole Instagram Accounts Simply by Asking Meta AI to Give Them Access
Hackers have been stealing Instagram accounts by asking Meta's AI support bot to link a target account to a new email address, without any verification needed. The AI sends a code to the attacker's email, which is then used to reset the password and gain full access. One victim only recovered his account after sending Meta's AI a link to a news article about the vulnerability.
Background
- The article reports that hackers stole Instagram accounts by simply asking Meta's AI-powered support chatbot to link a new email address to a target account, bypassing normal authentication. The AI bot sent a reset code to the attacker, granting full access. This was widely shared in security-researcher Telegram groups.<br>- Meta (formerly Facebook) has been aggressively deploying generative AI across its products under Mark Zuckerberg's leadership, including customer support chatbots that replace human agents. Critics argue this rollout prioritizes AI adoption over security and reliability.<br>- The author, John Gruber (Daring Fireball), recounts a friend with a rare three-letter Instagram username losing his account this way and only recovering it after sending the AI a link to the 404 Media article exposing the flaw — suggesting the AI itself then acknowledged the issue.<br>- 404 Media is an independent tech-journalism outlet that frequently covers cybersecurity, platform abuse, and the unintended consequences of AI systems. Their original reporting broke the story of this vulnerability.
Jimmy Wales announced that Wikipedia was live at wikipedia.com on January 15, 2001. The site was intended to be a "really quite snazzy" wiki complement to the Nupedia project, offering a more collaborative and less formal environment for building an encyclopedia.
Realta Fusion has achieved a milestone by generating electricity directly from a fusion reaction, reportedly a first. The breakthrough marks a significant step toward practical fusion energy, potentially offering a clean and nearly limitless power source.
OpenAI has announced Daybreak, a new initiative focused on advancing AI safety and alignment research to ensure artificial general intelligence benefits humanity.
SpaceX has announced plans to launch approximately one million satellites to create space-based data centres, according to the European Southern Observatory (ESO). The massive satellite constellation would significantly increase the number of objects in orbit, raising concerns about light pollution and interference with astronomical observations.
Ars Technica reports on the aftermath of the catastrophic failure of Blue Origin's New Glenn rocket, examining the investigation into the cause of the anomaly, the impact on the company's launch schedule and contracts, and the broader implications for the commercial space industry.